TGFp isoforms ((31, (32, and p3) are 25 kDa disulfide-linked homodimers that regulate cell proliferation, cell differentiation, and expression of extracellular matrix proteins. TGFps are best known for their tumor suppressor activity, and while loss of this activity leads to certain hereditary forms of colon and pancreatic cancer, in most cancers, including those of.the breast, the TGFp signaling pathway remains intact. This usually has adverse effects since the tumor promoting activities of TGFp, including its ability to stimulate immune suppression and endothelial-to-mesenchymal transitions, remain intact, yet its growth inhibitory activity is lost, due to either inactivation of the retinoblastoma gene product (pRB) or expression of growth stimulatory genes, such as c-myc. The three isoforms of TGFp share 71 - 79 % sequence identity and signal through a pair of structurally similar single-pass transmembrane receptors known as TRI and TRII. The isoforms nevertheless fufill distinct roles in vivo as shown by the non-overlapping and lethal phenotypes of the isoform specific -/- null mice, by differences in response induced by the addition of purified isoforms in tissue explant assays, and by the opposing roles they play in diseases, such as breast cancer. The origins of these differences are not understood, although based on previous cell-based crosslinking and structural studies, it might be due to differences in the manner by which they bind and assemble their receptors into a signaling complex. The first major objective of this proposal is to define the molecular architecture of the extracellular component of the TRI:TRII:TGFp signaling complex. This will provide fundamental information concerning the interdependent nature of TGFp receptor assembly, as well as how assembly differs for the isoforms whose monomers are fixed relative to one another (TGFps 1 and 2) compared to those whose monomers are not fixed (TGFpS). The second major component of this project concerns TGFp2, which also signals by binding and bringing together TRI and TRII, but which binds TRII weakly, and which is dependent upon a third cell surface TGFp binding protein, known as betaglycan, to induce its cellular responses. The objective of the proposed studies is to determine the structural basis underlying the mechanism by which the endoglin-like domain of the TGFp co-receptor betaglycan facilitates binding of TGFp2 to TRII. This will provide the first example of how one of the co-receptors in the family functions to selectively enhance the sensitivity of cells to a particular ligand isoform. The information derived from these studies will then be used to generate mutant TGFps to test whether differences among the isoforms in their manner of receptor binding indeed underlie their differences in biological activity using two different cell-based systems.